CN111774934A - Cutter health condition monitoring method, device and system based on end-to-end model - Google Patents
Cutter health condition monitoring method, device and system based on end-to-end model Download PDFInfo
- Publication number
- CN111774934A CN111774934A CN202010611581.8A CN202010611581A CN111774934A CN 111774934 A CN111774934 A CN 111774934A CN 202010611581 A CN202010611581 A CN 202010611581A CN 111774934 A CN111774934 A CN 111774934A
- Authority
- CN
- China
- Prior art keywords
- machine tool
- health condition
- real
- control signal
- time control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000036541 health Effects 0.000 title claims abstract description 114
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012544 monitoring process Methods 0.000 title claims abstract description 35
- 238000012806 monitoring device Methods 0.000 claims abstract description 40
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 239000013598 vector Substances 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 15
- 230000004913 activation Effects 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000003862 health status Effects 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 description 19
- 239000010410 layer Substances 0.000 description 16
- 230000003993 interaction Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000002349 favourable effect Effects 0.000 description 5
- 238000003754 machining Methods 0.000 description 5
- 238000003801 milling Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000002994 raw material Substances 0.000 description 4
- 238000005299 abrasion Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 229910001369 Brass Inorganic materials 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 239000010951 brass Substances 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000007514 turning Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
- B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
- B23Q17/0957—Detection of tool breakage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mechanical Engineering (AREA)
- Numerical Control (AREA)
Abstract
The invention relates to the technical field of cutter health condition monitoring, and particularly discloses an end-to-end model-based cutter health condition online monitoring method, which comprises the following steps: acquiring a real-time control signal of a machine tool spindle; preprocessing a real-time control signal of a machine tool spindle to obtain a preprocessed real-time control signal; sending the preprocessed real-time control signal to a master control slave device; receiving the health condition of the current cutter arranged on the main shaft of the machine tool fed back by the master control slave device; and sending the health condition of the current cutter arranged on the main shaft of the machine tool to an upper computer. The invention also discloses an end-to-end model-based cutter health condition online monitoring device and system. The cutter health condition online monitoring method based on the end-to-end model can automatically monitor the wear degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency.
Description
Technical Field
The invention relates to the technical field of cutter health condition monitoring, in particular to a cutter health condition online monitoring method, device and system based on an end-to-end model.
Background
With the economic development, the domestic industry and the manufacturing industry become more and more important, and industrial equipment tends to be more and more intelligent and integrated. The machine tool is used as a basic production tool of a manufacturing factory and is an important and expensive production cost, and a machining tool bit used by the machine tool is a consumable product for ensuring the product quality in the production and machining process, and the tool bit must be ensured to be in a normal state in the actual production process. However, the actual service life of the tool fluctuates within a certain range due to the different conditions of the production environment and the processing environment, and the time for replacing the tool is difficult to grasp. If the cutter is aged or the cutter is not replaced due to the wear of the broken cutter, the machining is continued, so that the machining raw materials are damaged, and the machine tool is seriously damaged even; if the tool is replaced in advance while the tool is still in use, this reduces the utility of the tool and virtually increases the cost. The method can reduce the cost of the cutter, prolong the service life of the machine tool and save raw materials. The method is particularly important for manufacturing enterprises to reduce the manufacturing cost and improve the competitiveness.
At present, most factories adopt a counting method or an intermittent monitoring method to realize the health condition evaluation of the cutter, two strategies are mainly adopted for replacing the cutter, one is a cutter cutting method for uniformly setting the service time of the cutter by the factory, the service time of the cutter is recorded, the cutter is replaced after the service time reaches the upper limit of the service time, and whether the cutter can be used or not is judged, the method is simple and convenient to implement, but the consumption ratio of the cutter and raw materials is large, if the replaced cutter can be used continuously, the potential value of the cutter is wasted, the production cost is improved, if the replaced cutter is worn out too early to be used continuously, some products machined before the cutter cannot reach the precision required by actual production, the raw materials are wasted, a machine tool can be damaged, and the production cost is improved. The other method is that on the former method, experienced workers are matched, and for the experienced workers, whether the cutter can not be used any more can be judged through the processing sound of the cutter, and for the cutter with serious abrasion, the cutter can be distinguished and replaced in time, but the method can not judge the residual service time of the cutter, and for workers in factories, due to the existence of subjective judgment factors, the judgment cannot be completely correct, and the consumed manpower is relatively large.
Disclosure of Invention
The invention provides a method, a device and a system for monitoring the health condition of a cutter on line based on an end-to-end model, which solve the problem that the health condition of the cutter can not be effectively monitored in the related technology.
As a first aspect of the present invention, there is provided an online tool health monitoring method based on an end-to-end model, comprising:
acquiring a real-time control signal of a machine tool spindle, wherein the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
sending the preprocessed real-time control signal to a master slave device, wherein the master slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
receiving the health condition of the current cutter arranged on the main shaft of the machine tool fed back by the master control slave device;
and sending the health condition of the current cutter arranged on the main shaft of the machine tool to an upper computer.
Further, after the step of obtaining the real-time control signal of the spindle of the machine tool, the method further comprises the following steps:
and sending the real-time control signal of the machine tool spindle to an upper computer.
Further, the expression of the end-to-end tool model identification algorithm model is as follows:
wherein,
representing the output of the end-to-end tool model identification algorithm model, conv representing the reel base layer, and I representing the vector.
Further, representing the weight by an m-dimensional vector H, wherein the vectors H corresponding to different channels are different, representing data to be convolved, which are obtained by calculating original data, by a k-dimensional vector I, representing a bias constant by b, and taking the ith sub-vector of I in the ith convolution operation with the step length d
Convolving with a weight vector H to obtain:
wherein HjRepresents the jth element of the weight vector H; the output data S is obtained by the ReLU activation function:
Ui＝ReLU(S(i))＝max(0,S(i))，
obtain an output
The output U passes through the pooling layer pool to obtain:
wherein, p represents the pooling size, e represents the pooling step length, and after passing through the convolution layer and the pooling layer, the output data is transmitted to the flat layer Flatt and is output as a one-dimensional vector F ═ F1,F2,...,Fq]And q represents the length of the output vector.
Further, the full-link layer FCN includes an activation function ReLU and a matrix dot product calculation, and the expression of the activation function ReLU and the matrix dot product calculation is as follows:
O＝ReLU(W·F)，
wherein W represents a weight in the FCN layer, O ═ O1,O2,...,ON]N represents the total health condition category number of the cutter, the output activation function act is a softmax function, and the end-to-end cutter model identification algorithm model is output
Expressed as:
as another aspect of the invention, a tool health online monitoring device based on an end-to-end model is provided, which includes:
the system comprises an acquisition module, a processing module and a control module, wherein the acquisition module is used for acquiring a real-time control signal of a machine tool spindle, and the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
the preprocessing module is used for preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
the first sending module is used for sending the preprocessed real-time control signal to a master-slave device, and the master-slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
the receiving module is used for receiving the health condition of the current cutter arranged on the main shaft of the machine tool and fed back by the master control slave device;
and the second sending module is used for sending the health condition of the current cutter arranged on the machine tool spindle to the upper computer.
As another aspect of the present invention, an end-to-end model-based online monitoring system for health status of a tool is provided, wherein the system comprises an acquisition device, a master slave device, an upper computer and the end-to-end model-based online monitoring device for health status of a tool, and the acquisition device, the master slave device and the upper computer are all in communication connection with the end-to-end model-based online monitoring device for health status of a tool;
the acquisition device is used for acquiring real-time control signals of the machine tool spindle;
the end-to-end model-based tool health condition online monitoring device is used for preprocessing a real-time control signal of the machine tool spindle and sending the preprocessed real-time control signal to the master control slave device;
the master control slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of the current tool arranged on the main shaft of the machine tool;
the end-to-end model-based tool health condition online monitoring device is also used for receiving the health condition of the current tool on the machine tool spindle and sending the health condition of the current tool on the machine tool spindle to an upper computer;
and the upper computer is used for receiving and displaying the health condition of the current cutter on the machine tool spindle.
Further, the end-to-end model-based tool health condition online monitoring device is further used for sending the real-time control signal of the machine tool spindle to the upper computer, and the upper computer is used for storing and backing up the real-time control signal of the machine tool spindle.
Further, the acquisition device comprises a non-contact sensor.
Further, the cutter health condition online monitoring device based on the end-to-end model comprises an embedded processor chip with a core of cotex-M4.
According to the method for monitoring the health condition of the cutter on line based on the end-to-end model, the health condition of the cutter is obtained by acquiring the real-time control signal of the machine tool spindle, preprocessing the real-time control signal of the machine tool spindle and then sending the preprocessed real-time control signal to the master-control slave device for calculation and analysis of the health condition of the cutter. The cutter health condition on-line monitoring method based on the end-to-end model can automatically monitor the wear degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency. Through the accurate cutter degree of wearing and tearing of distinguishing, can change the wearing and tearing cutter at more accurate time point, be favorable to more that the manufacturing plant saves the processing cost, improves product competitiveness.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
FIG. 1 is a flow chart of the method for monitoring the health of a tool on line based on an end-to-end model according to the present invention.
Fig. 2 is a schematic view of an interaction structure of the tool health online monitoring device based on the end-to-end model and the master-slave device provided by the invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged under appropriate circumstances in order to facilitate the description of the embodiments of the invention herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In this embodiment, an end-to-end model-based online monitoring method for health condition of a tool is provided, and fig. 1 is a flowchart of an end-to-end model-based online monitoring method for health condition of a tool according to an embodiment of the present invention, as shown in fig. 1, including:
s110, acquiring a real-time control signal of a machine tool spindle, wherein the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
s120, preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
s130, sending the preprocessed real-time control signal to a master slave device, wherein the master slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear recognition algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
s140, receiving the health condition of the current cutter arranged on the main shaft of the machine tool and fed back by the master control slave device;
and S150, sending the health condition of the current cutter arranged on the main shaft of the machine tool to an upper computer.
According to the method for monitoring the health condition of the cutter on line based on the end-to-end model, provided by the embodiment of the invention, the health condition of the cutter is obtained by acquiring the real-time control signal of the machine tool spindle, preprocessing the real-time control signal of the machine tool spindle and then sending the preprocessed real-time control signal to the master-control slave device for calculation and analysis of the health condition of the cutter. The cutter health condition on-line monitoring method based on the end-to-end model can automatically monitor the wear degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency. Through the accurate cutter degree of wearing and tearing of distinguishing, can change the wearing and tearing cutter at more accurate time point, be favorable to more that the manufacturing plant saves the processing cost, improves product competitiveness.
It should be understood that the calculation and analysis efficiency can be effectively improved by the implementation mode that the evaluation and acquisition of the health condition of the current cutter are realized on the master control slave device according to the end-to-end cutter wear identification algorithm model, and the result is fed back to the end-to-end model-based online cutter health condition monitoring device.
Specifically, the method further includes, after the step of obtaining a real-time control signal of the spindle of the machine tool:
and sending the real-time control signal of the machine tool spindle to an upper computer.
Specifically, the expression of the end-to-end tool model identification algorithm model is as follows:
wherein,
representing the output of the end-to-end tool model identification algorithm model, conv representing the reel base layer, and I representing the vector.
Further, representing the weight by an m-dimensional vector H, wherein the vectors H corresponding to different channels are different, representing data to be convolved, which are obtained by calculating original data, by a k-dimensional vector I, representing a bias constant by b, and taking the ith sub-vector of I in the ith convolution operation with the step length d
Convolving with a weight vector H to obtain:
wherein HjRepresents the jth element of the weight vector H; the output data S is obtained by the ReLU activation function:
Ui＝ReLU(S(i))＝max(0,S(i))，
obtain an output
The output U passes through the pooling layer pool to obtain:
wherein, p represents the pooling size, e represents the pooling step length, and after passing through the convolution layer and the pooling layer, the output data is transmitted to the flat layer Flatt and is output as a one-dimensional vector F ═ F1,F2,...,Fq]And q represents the length of the output vector.
Further, the full-link layer FCN includes an activation function ReLU and a matrix dot product calculation, and the expression of the activation function ReLU and the matrix dot product calculation is as follows:
O＝ReLU(W·F)，
wherein W represents a weight in the FCN layer, O ═ O1,O2,...,ON]N represents the total health condition category number of the cutter, the output activation function act is a softmax function, and the end-to-end cutter model identification algorithm model is output
Expressed as:
it should be understood that for the above output
In which the individual layers have the form of successive layers or iterations, by comparing the output vectors
And the internal value can judge the abrasion state of the cutter.
It should be noted that, in the process of processing the workpieces one by the numerical control machine, the moving track of the tool is in regular cyclic reciprocating motion, so the control current of the machine tool spindle is a periodic signal, the control current can be converted into a smaller voltage signal through a non-contact sensor (such as a hall sensor), so that the purpose of obtaining a machine tool signal without modifying the machine tool is achieved, and other signals such as a machine tool vibration signal can be obtained by a similar method. The signals are further acquired by a machine tool spindle data acquisition device of the tool health condition online monitoring terminal and converted into data streams in the tool health condition online monitoring device based on the end-to-end model, an internet communication device and an operation platform device are arranged on the tool health condition online monitoring device based on the end-to-end model, and a scheduling mode with a mu cos operation system as a main body is used inside the tool health condition online monitoring device based on the end-to-end model. In this embodiment, the internet communication device mainly uses a wired network card, works in a local area network environment adapted to a factory production environment, and is collocated with a lightweight internet protocol stack Lwip. The operation platform device mainly comprises display equipment and worker operation equipment, and a human-computer interaction interface program and required resources are scheduled by an operation system in the main control, so that a convenient human-computer interaction control effect can be achieved. In addition, a simple server is built in the equipment, the effect of a software operation platform corresponding to the function of the hardware operation platform device can be achieved by matching with the Internet equipment, and a worker can use the network remote control terminal equipment. In practice, the worker may interact with either or both of the platforms simultaneously to control the device.
The main process of inputting the preprocessed real-time control signal into the tool wear identification algorithm model is as follows: three-phase current signals of a machine tool spindle are converted into data streams representing current after passing through a sensor acquisition device and a cutter health condition online monitoring device based on an end-to-end model, the data streams are processed through segmentation and the like and then input into the end-to-end cutter wear identification algorithm model as original data, and identification data types are output so as to distinguish the cutter state.
As another embodiment of the present invention, an on-line tool health monitoring device based on an end-to-end model is provided, which includes:
the system comprises an acquisition module, a processing module and a control module, wherein the acquisition module is used for acquiring a real-time control signal of a machine tool spindle, and the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
the preprocessing module is used for preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
the first sending module is used for sending the preprocessed real-time control signal to a master-slave device, and the master-slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
the receiving module is used for receiving the health condition of the current cutter arranged on the main shaft of the machine tool and fed back by the master control slave device;
and the second sending module is used for sending the health condition of the current cutter arranged on the machine tool spindle to the upper computer.
According to the end-to-end model-based online monitoring device for the health condition of the cutter, provided by the embodiment of the invention, the health condition of the cutter is obtained by acquiring the real-time control signal of the machine tool spindle, preprocessing the real-time control signal of the machine tool spindle and then sending the preprocessed real-time control signal to the master-control slave device for calculation and analysis of the health condition of the cutter. The cutter health condition on-line monitoring method based on the end-to-end model can automatically monitor the wear degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency. Through the accurate cutter degree of wearing and tearing of distinguishing, can change the wearing and tearing cutter at more accurate time point, be favorable to more that the manufacturing plant saves the processing cost, improves product competitiveness.
As another embodiment of the present invention, an end-to-end model-based online monitoring system for health status of a tool is provided, wherein the system comprises an acquisition device, a master slave device, an upper computer, and the end-to-end model-based online monitoring device for health status of a tool, and the acquisition device, the master slave device, and the upper computer are all in communication connection with the end-to-end model-based online monitoring device for health status of a tool;
the acquisition device is used for acquiring real-time control signals of the machine tool spindle;
the end-to-end model-based tool health condition online monitoring device is used for preprocessing a real-time control signal of the machine tool spindle and sending the preprocessed real-time control signal to the master control slave device;
the master control slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of the current tool arranged on the main shaft of the machine tool;
the end-to-end model-based tool health condition online monitoring device is also used for receiving the health condition of the current tool on the machine tool spindle and sending the health condition of the current tool on the machine tool spindle to an upper computer;
and the upper computer is used for receiving and displaying the health condition of the current cutter on the machine tool spindle.
The system for on-line monitoring the health condition of the cutter based on the end-to-end model, provided by the embodiment of the invention, adopts the cutter health condition on-line monitoring device based on the end-to-end model, and obtains the health condition of the cutter by acquiring the real-time control signal of the machine tool spindle, preprocessing the real-time control signal of the machine tool spindle and then sending the preprocessed real-time control signal to the master-control slave device for calculation and analysis of the health condition of the cutter. The cutter health condition on-line monitoring method based on the end-to-end model can automatically monitor the wear degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency. Through the accurate cutter degree of wearing and tearing of distinguishing, can change the wearing and tearing cutter at more accurate time point, be favorable to more that the manufacturing plant saves the processing cost, improves product competitiveness.
Specifically, the end-to-end model-based tool health condition online monitoring device is further used for sending the real-time control signal of the machine tool spindle to the upper computer, and the upper computer is used for storing and backing up the real-time control signal of the machine tool spindle.
Preferably, the acquisition device comprises a non-contact sensor.
Preferably, the tool health online monitoring device based on the end-to-end model comprises an embedded processor chip with a core of cotex-M4.
Specifically, according to the end-to-end model-based tool health condition online monitoring system provided by the embodiment of the invention, a friend-and-good vertical machining center machine tool is used for grooving and milling a common square brass workpiece, a 4-blade straight milling cutter with the diameter of 1mm is used as the tool, the rotating speed is about 6000 revolutions, and the machining track of the tool is in regular cyclic reciprocating motion, so that the control current of a machine tool spindle is also a periodic signal, and the period can be in units of seconds. The control current can be converted into a small weak current voltage signal through a non-contact sensor, for example, a Hall current sensor of a Honeywell can be used for collecting the main shaft control current, and a machine tool motor is driven by three-phase current, so that 3 current sensors are needed to obtain 3 signals of the control current respectively, so that the internal signals of the machine tool can be obtained without changing the machine tool, and other signals such as machine tool vibration signals can also be obtained by a similar method. The machine tool related signals are further acquired by a machine tool spindle data acquisition device of the tool health condition online monitoring terminal and converted into data streams in the tool health condition online monitoring device based on an end-to-end model, and a data acquisition device body adopts an AD7606 chip and can acquire data at a sampling rate of 200kSPS through 8 channels. The cutter health condition on-line monitoring device based on the end-to-end model is used as a control center of a terminal device, certain computing capacity and regulation capacity are needed, for example, an embedded processor chip with a core of cotex-M4 is used as a main control chip, a data acquisition device, an internet communication device and an operation platform device are connected, the master frequency of the chip can reach more than 100M, and meanwhile, a mu cos-II operating system is used as an interface of internal scheduling resources. The Internet communication device and the operation platform device are both arranged on the cutter health condition online monitoring device based on the end-to-end model, the communication device uses the LAN8720 network card chip, the system can reach the Internet communication speed of a hundred mega network, the matched network protocol stack is a lightweight embedded protocol stack LWIP, and the communication is more efficient by shortening the interlayer communication mode.
Based on the purpose of rapid communication, a UDP protocol can be used for transmitting data streams, and an HTTP server is established in the terminal so as to be capable of remotely controlling the terminal, so that all terminal equipment connected with the upper computer can be managed and controlled by only one upper computer. The operation platform device comprises a display device and a worker operation device, an operation system in the cutter health condition on-line monitoring device based on the end-to-end model schedules resources such as screens and buttons used by the operation platform and a man-machine interaction interface program, so that a hardware terminal can be conveniently controlled, the display device uses a 4.3-inch TFTLCD screen for displaying menus of the operation interface and various data of the terminal, workers can conveniently control the terminal and acquire states of a machine tool cutter and the terminal device according to prompts in the interface, the operation device comprises various keys and state lamps, and interaction with the equipment can be conveniently and rapidly carried out in cooperation with menu prompts. Besides the hardware interaction device, the terminal equipment is internally provided with an HTTP server for software interaction, an upper computer which is connected with the terminal through a network can monitor and manage the equipment through a browser webpage, and the effect of a software operation platform corresponding to the function of the hardware operation platform device is achieved, so that workers can also use the network to remotely control the terminal equipment. In actual operation, a worker can interact with the terminal equipment for one of the two platforms simultaneously or optionally, so as to uniformly control and manage all the terminal equipment.
Data streams of current signals in the cutter health condition online monitoring device based on the end-to-end model are divided into flow directions according to different purposes of use, wherein one of the flow directions is that original data streams can be directly transmitted to a control upper computer through a UDP (user Datagram protocol) and a network card after being packaged into network frames in a main control without signal processing, and the data streams are all stored by the upper computer and can be used as backup records of data and source data of other later-stage processing; the other flow direction is data analysis and processing in parallel, because the tool wear identification algorithm can be called on the master control slave device controlled by the tool health condition on-line monitoring device based on the end-to-end model, the data flow to be processed can be simply preprocessed by the tool health condition on-line monitoring device based on the end-to-end model and then transferred to the master control slave device which is connected with the tool health condition on-line monitoring device based on the end-to-end model and managed by the tool health condition on-line monitoring device based on the end-to-end model, the tool wear identification algorithm in the master control slave device obtains the wear condition of the tool, in the embodiment, the master control slave device is an artificial intelligent chip, realizes the tool wear identification algorithm, after the tool health condition on-line monitoring device based on the end-to-end model receives the result sent by the slave master control device, thereby determining whether to send a tool change signal to the machine tool. The signal flow of the system as a whole can be seen in fig. 2.
In summary, the method, the device and the system for monitoring the health condition of the cutter on line based on the end-to-end model provided by the invention utilize intelligent equipment to monitor the working condition of the machine tool in real time, analyze the wear condition of the cutter by using big data, and replace the cutter in time when the cutter can not be used continuously, thereby overcoming the defects of the prior art. The mode can automatically monitor the abrasion degree of the cutter in real time, and solves the problems that the existing scheme needs a large amount of manpower and is low in efficiency. Through the accurate cutter degree of wearing and tearing of distinguishing, can change the wearing and tearing cutter at more accurate time point, be favorable to more that the manufacturing plant saves the processing cost, improves product competitiveness. The scheme of the invention can reduce the cost by 15 percent and improve the production efficiency by 20 percent (the number is related to a specific processing technology), and the diameter of the detected cutter can reach 1mm at least. The specific application scenes comprise mobile phone shell milling, automobile part turbine shaft turning, small-load milling of aerospace heat dissipation parts and the like. The method can be theoretically applied to the health condition evaluation links of other industrial production tools similar to cutters.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (10)
1. An end-to-end model-based tool health condition online monitoring method is characterized by comprising the following steps:
acquiring a real-time control signal of a machine tool spindle, wherein the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
sending the preprocessed real-time control signal to a master slave device, wherein the master slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
receiving the health condition of the current cutter arranged on the main shaft of the machine tool fed back by the master control slave device;
and sending the health condition of the current cutter arranged on the main shaft of the machine tool to an upper computer.
2. The method for on-line monitoring health status of tool based on end-to-end model according to claim 1, further comprising after the step of obtaining real-time control signal of machine tool spindle:
and sending the real-time control signal of the machine tool spindle to an upper computer.
3. The on-line tool health monitoring method based on the end-to-end model according to claim 1, wherein the end-to-end tool model recognition algorithm model is represented by the following formula:
wherein,
representing the output of the end-to-end tool model identification algorithm model, conv representing the reel base layer, and I representing the vector.
4. The on-line tool health monitoring method according to claim 3, wherein the weight is represented by m-dimensional vector H, the vectors H corresponding to different channels are different, the data to be convolved calculated from the original data is represented by k-dimensional vector I, the bias constant is represented by b, and in the ith convolution operation with step d, the ith subvector of I is taken
Convolving with a weight vector H to obtain:
wherein HjRepresents the jth element of the weight vector H; the output data S is obtained by the ReLU activation function:
Ui＝ReLU(S(i))＝max(0,S(i))，
obtain an output
The output U passes through the pooling layer pool to obtain:
wherein, p represents the pooling size, e represents the pooling step length, and after passing through the convolution layer and the pooling layer, the output data is transmitted to the flat layer Flatt and is output as a one-dimensional vector F ═ F1,F2,...,Fq]And q represents the length of the output vector.
5. The on-line tool health monitoring method based on the end-to-end model according to claim 4, wherein the full-link layer FCN comprises an activation function ReLU and a matrix dot-product calculation, and the expression of the activation function ReLU and the matrix dot-product calculation is as follows:
O＝ReLU(W·F)，
wherein W represents a weight in the FCN layer, O ═ O1,O2,...,ON]N represents the total health condition category number of the cutter, the output activation function act is a softmax function, and the end-to-end cutter model identification algorithm model is output
Expressed as:
6. an end-to-end model-based tool health online monitoring device, comprising:
the system comprises an acquisition module, a processing module and a control module, wherein the acquisition module is used for acquiring a real-time control signal of a machine tool spindle, and the real-time control signal of the machine tool spindle comprises a voltage signal corresponding to a control current signal of the machine tool spindle;
the preprocessing module is used for preprocessing the real-time control signal of the machine tool spindle to obtain a preprocessed real-time control signal;
the first sending module is used for sending the preprocessed real-time control signal to a master-slave device, and the master-slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of a current tool arranged on a main shaft of the machine tool;
the receiving module is used for receiving the health condition of the current cutter arranged on the main shaft of the machine tool and fed back by the master control slave device;
and the second sending module is used for sending the health condition of the current cutter arranged on the machine tool spindle to the upper computer.
7. An end-to-end model-based online monitoring system for the health condition of a cutter is characterized by comprising an acquisition device, a master slave device, an upper computer and the end-to-end model-based online monitoring device for the health condition of the cutter as claimed in claim 6, wherein the acquisition device, the master slave device and the upper computer are in communication connection with the end-to-end model-based online monitoring device for the health condition of the cutter;
the acquisition device is used for acquiring real-time control signals of the machine tool spindle;
the end-to-end model-based tool health condition online monitoring device is used for preprocessing a real-time control signal of the machine tool spindle and sending the preprocessed real-time control signal to the master control slave device;
the master control slave device is used for inputting the preprocessed real-time control signal into an end-to-end tool wear identification algorithm model to obtain the health condition of the current tool arranged on the main shaft of the machine tool;
the end-to-end model-based tool health condition online monitoring device is also used for receiving the health condition of the current tool on the machine tool spindle and sending the health condition of the current tool on the machine tool spindle to an upper computer;
and the upper computer is used for receiving and displaying the health condition of the current cutter on the machine tool spindle.
8. The end-to-end model-based tool health online monitoring system according to claim 7, wherein the end-to-end model-based tool health online monitoring device is further configured to send a real-time control signal of the machine tool spindle to the upper computer, and the upper computer is configured to store and backup the real-time control signal of the machine tool spindle.
9. The end-to-end model-based online tool health monitoring system of claim 7, wherein the collection device comprises a non-contact sensor.
10. The end-to-end model-based online tool health monitoring system of claim 7, wherein the end-to-end model-based online tool health monitoring device comprises an embedded processor chip with a core of cotex-M4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010611581.8A CN111774934A (en) | 2020-06-30 | 2020-06-30 | Cutter health condition monitoring method, device and system based on end-to-end model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010611581.8A CN111774934A (en) | 2020-06-30 | 2020-06-30 | Cutter health condition monitoring method, device and system based on end-to-end model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111774934A true CN111774934A (en) | 2020-10-16 |
Family
ID=72760745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010611581.8A Pending CN111774934A (en) | 2020-06-30 | 2020-06-30 | Cutter health condition monitoring method, device and system based on end-to-end model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111774934A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5464780A (en) * | 1977-11-02 | 1979-05-24 | Toshiba Mach Co Ltd | Unattended operating system in machining |
| CN101209530A (en) * | 2006-12-26 | 2008-07-02 | 财团法人工业技术研究院 | Device and method for detecting cutting tool state |
| CN105234746A (en) * | 2015-11-25 | 2016-01-13 | 苏州多荣自动化科技有限公司 | Online tool wear monitoring system and detection method thereof |
| CN106514434A (en) * | 2016-12-30 | 2017-03-22 | 华中科技大学 | Milling tool abrasion monitoring method based on data |
| CN106779073A (en) * | 2016-12-27 | 2017-05-31 | 西安石油大学 | Media information sorting technique and device based on deep neural network |
| CN108873813A (en) * | 2018-06-25 | 2018-11-23 | 山东大学 | Tool wear degree detection method based on main shaft of numerical control machine tool servo motor current signal |
| JP2019207576A (en) * | 2018-05-29 | 2019-12-05 | ファナック株式会社 | Life prediction device and mechanical learning device |
-
2020
- 2020-06-30 CN CN202010611581.8A patent/CN111774934A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5464780A (en) * | 1977-11-02 | 1979-05-24 | Toshiba Mach Co Ltd | Unattended operating system in machining |
| CN101209530A (en) * | 2006-12-26 | 2008-07-02 | 财团法人工业技术研究院 | Device and method for detecting cutting tool state |
| CN105234746A (en) * | 2015-11-25 | 2016-01-13 | 苏州多荣自动化科技有限公司 | Online tool wear monitoring system and detection method thereof |
| CN106779073A (en) * | 2016-12-27 | 2017-05-31 | 西安石油大学 | Media information sorting technique and device based on deep neural network |
| CN106514434A (en) * | 2016-12-30 | 2017-03-22 | 华中科技大学 | Milling tool abrasion monitoring method based on data |
| JP2019207576A (en) * | 2018-05-29 | 2019-12-05 | ファナック株式会社 | Life prediction device and mechanical learning device |
| CN108873813A (en) * | 2018-06-25 | 2018-11-23 | 山东大学 | Tool wear degree detection method based on main shaft of numerical control machine tool servo motor current signal |
Non-Patent Citations (1)
| Title |
|---|
| 戴稳: "基于深度学习的铣刀磨损状态识别及预测方法研究", 《中国优秀硕士学位论文全文数工程科技Ⅰ辑据库》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103823409B (en) | Digit Control Machine Tool machining state multiparameter online actively monitoring system and its implementation | |
| CN203894596U (en) | Multi-parameter online active monitoring system for machining states of numerical control machine bed | |
| CN101334656B (en) | Numerical control machine processability monitoring system | |
| CN104808585B (en) | A kind of quick inspection method of lathe health status | |
| KR101754721B1 (en) | A System and Apparatus that identifies, captures, classifies and deploys tribal knowledge unique to each operator in a semi-automated manufacturing set-up to execute Automatic Technical Superintending Operations to improve manufacturing system performance and the method/s therefor | |
| CN114089690B (en) | Edge computing device and method for workshop monitoring | |
| CN103760820A (en) | Evaluation device of state information of machining process of numerical control milling machine | |
| TWI618018B (en) | Production mangement method and system of utilizing power consumption feature | |
| CN111571302B (en) | Cutting fluid amount adjusting device and system | |
| US7409261B2 (en) | Data management and networking system and method | |
| CN110297464A (en) | Failure prediction system and its method based on sensing data on numerically-controlled machine tool | |
| CN111774932B (en) | Cutter health condition online monitoring method, device and system | |
| CN110210366A (en) | Assembling and screwing process sample acquisition system, deep learning network and monitoring system | |
| CN111352398A (en) | Intelligent precision machining unit | |
| CN117170310A (en) | Digital twin-based remote fault diagnosis method and system for numerical control machine tool | |
| CN115309106A (en) | Machine tool thermal field simulation system based on multi-source heterogeneous data | |
| Kubota et al. | A cyber-physical machine tool framework based on STEP-NC | |
| CN117991752B (en) | Equipment fault prediction system and method based on digital twin and Internet of things technology | |
| CN110488775A (en) | Equipment state judgement and yield beat statistical system and method | |
| CN214067660U (en) | Monitoring system based on Internet of things | |
| KR20230032675A (en) | System for collecting data using computerized numerical control mother machine | |
| CN116141058B (en) | Self-adaptive matching and control method, system and equipment for numerical control machine tool | |
| CN111774934A (en) | Cutter health condition monitoring method, device and system based on end-to-end model | |
| CN101644922A (en) | Digital centralized intelligent monitoring system for machine tool | |
| CN111598251B (en) | CNC predictive maintenance system and method based on machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221109 Address after: 226010 Floor 4, Building 8, Zilang Science and Technology City, No. 60 Chongzhou Avenue, Xiaohai Street, Nantong Development Zone, Jiangsu Province Applicant after: Yuanshi Intelligent Technology (Nantong) Co.,Ltd. Address before: 214174 329 Weir Road, Huishan District, Wuxi, Jiangsu Applicant before: HUST-WUXI Research Institute |
|
| TA01 | Transfer of patent application right |